INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

CAM

-0.13

 kazan

-0.13

evice

-0.13

usb

-0.13

.CASCADE

-0.13

cam

-0.13

 bondage

-0.13

asma

-0.13

 labore

-0.13

ÄĽr

-0.13

POSITIVE LOGITS

 gang

0.16

 beat

0.15

 criminal

0.15

lul

0.14

gang

0.14

iminal

0.14

 attack

0.14

Dag

0.14

kowski

0.14

downloads

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.